Statistical Prediction of ENSO from Subsurface Sea Temperature Using a Nonlinear Dimensionality Reduction

Numerous statistical and dynamical models have been developed in recent years to forecast ENSO events. However, for most of these models predictability for lead times over 10 months is limited. It has been hypothesized that the tropical Pacific thermocline structure may have critical information to permit longer lead ENSO forecasts. Models that use subsurface sea temperature information have already been known to produce better long lead forecasts. Here, a two-stage statistical ENSO forecasting model is developed and demonstrated using the spatially distributed depth of the 20°C isotherm (D20) as a proxy for the thermocline. In the first stage, a nonlinear dimension reduction method [maximum variance unfolding (MVU)] is used to decompose the D20 data into canonical modes. The leading spatial patterns as well as lagged values of Niño-3 are then used as predictors in a set of linear regression models to predict the Niño-3 index at lead times of up to 24 months. Cross-validated forecasts using this methodology are shown to have higher skill than those that use a dimension reduction of the same thermocline data using principal component analysis (PCA). The first three modes of the D20 data as revealed by MVU account for 89% of the variance of the data, as compared to only 48% of the variance if PCA is used. The spatial patterns of the MVU modes partition the data field in a different way than the PC modes, even though some similarities exist as to the main regions that are active. These patterns and their temporal structure are discussed here, with a view to understanding the possible source of the longer-range predictability of ENSO using the MVU modes. The skill of the PCA- and the MVU-based forecasts of Niño-3 varies depending on the starting month of the forecast for short lead times (5–10 months). However, for the lead times longer than 1 yr, the MVU-based forecast skill is not seasonally variable, while the PCA-based models do not provide significant skill at these lead times irrespective of the starting month of the forecast. Similar conclusions are obtained for forecast models for the Niño-3.4 and Niño-1.2 indices. The differences between the MVU- and PCA-based models are most marked for the Niño-1.2 long lead forecasts.

[1]  Ashish Sharma,et al.  Integrating Ocean Subsurface Temperatures in Statistical ENSO Forecasts , 2005 .

[2]  Fei-Fei Jin,et al.  An Equatorial Ocean Recharge Paradigm for ENSO. Part I: Conceptual Model , 1997 .

[3]  A. Barnston,et al.  Predictive Skill of Statistical and Dynamical Climate Models in SST Forecasts during the 1997-98 El Niño Episode and the 1998 La Niña Onset. , 1999 .

[4]  F. Schwing,et al.  A new climate regime in northeast pacific ecosystems , 2003 .

[5]  Kilian Q. Weinberger,et al.  Learning a kernel matrix for nonlinear dimensionality reduction , 2004, ICML.

[6]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[7]  D Ph,et al.  Conceptual Model , 2017, Encyclopedia of GIS.

[8]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[9]  Ming Ji,et al.  An Improved Coupled Model for ENSO Prediction and Implications for Ocean Initialization. Part I: The Ocean Data Assimilation System , 1998 .

[10]  L. Goddarda,et al.  CURRENT APPROACHES TO SEASONAL-TO-INTERANNUAL CLIMATE PREDICTIONS , 2000 .

[11]  Stanley B. Goldenberg,et al.  Documentation o f a Highly ENSO-Related SST Région in thé Equatorial Pacific , 1997 .

[12]  J. Michaelsen.  Cross-Validation in Statistical Climate Forecast Models , 1987 .

[13]  Michael J. McPhaden,et al.  Observations of Warm Water Volume Changes in the Equatorial Pacific and Their Relationship to El Niño and La Niña , 2000 .

[14]  Balaji Rajagopalan,et al.  Analyses of global sea surface temperature 1856–1991 , 1998 .

[15]  Ming Ji,et al.  ENSO Prediction with Markov Models: The Impact of Sea Level , 2000 .

[16]  C. Ropelewski,et al.  Current approaches to seasonal to interannual climate predictions , 2001 .

[17]  Michael J. McPhaden,et al.  Tropical Pacific Ocean heat content variations and ENSO persistence barriers , 2003 .

[18]  Allan J. Clarke,et al.  Improving El Niño prediction using a space‐time integration of Indo‐Pacific winds and equatorial Pacific upper ocean heat content , 2003 .

[19]  M. Cane,et al.  A Model El Niñ–Southern Oscillation , 1987 .

[20]  Fei-Fei Jin,et al.  An Equatorial Ocean Recharge Paradigm for ENSO. Part II: A Stripped-Down Coupled Model , 1997 .

[21]  B. Borchers.  CSDP, A C library for semidefinite programming , 1999 .

[22]  Mikhail Belkin,et al.  Laplacian Eigenmaps for Dimensionality Reduction and Data Representation , 2003, Neural Computation.

[23]  W. Drosdowsky.  Statistical prediction of ENSO (Nino 3) using sub‐surface temperature data , 2006 .

[24]  Kilian Q. Weinberger,et al.  Unsupervised Learning of Image Manifolds by Semidefinite Programming , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[25]  Peter J. Webster,et al.  Monsoon and Enso: Selectively Interactive Systems , 1992 .

[26]  C. Ropelewski,et al.  Global and Regional Scale Precipitation Patterns Associated with the El Niño/Southern Oscillation , 1987 .

[27]  Thomas M. Smith,et al.  Improved Global Sea Surface Temperature Analyses Using Optimum Interpolation , 1994 .

[28]  David L. T. Anderson,et al.  Decadal and Seasonal Dependence of ENSO Prediction Skill , 1995 .

[29]  T. Smith,et al.  Ocean Model Response to Temperature Data Assimilation and Varying Surface Wind Stress: Intercomparisons and Implications for Climate Forecast , 1995 .

[30]  R. E. Livezey,et al.  Statistical Field Significance and its Determination by Monte Carlo Techniques , 1983 .

[31]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[32]  K. Hanawa,et al.  Two different features of discharge of equatorial upper ocean heat content related to El Niño events , 2006 .

[33]  Francisco P. Chavez,et al.  From Anchovies to Sardines and Back: Multidecadal Change in the Pacific Ocean , 2003, Science.

[34]  Dongxiao Zhang,et al.  Pacific Ocean circulation rebounds , 2004 .

[35]  T. Smith,et al.  The Impact of Pacific Ocean Subsurface Data on Operational Prediction of Tropical Pacific SST at the NCEP , 1995 .

[36]  Tony Jebara,et al.  Minimum Volume Embedding , 2007, AISTATS.

[37]  Joshua B. Tenenbaum,et al.  Mapping a Manifold of Perceptual Observations , 1997, NIPS.

[38]  S. Mason,et al.  Probabilistic Precipitation Anomalies Associated with ENSO , 2001 .

[39]  K. Wyrtki.  Fluctuations of the Dynamic Topography in the Pacific Ocean , 1975 .

[40]  Alexey Kaplan,et al.  Predictability of El Niño over the past 148 years , 2004, Nature.

[41]  Katharine M. Kanak,et al.  Currents of Change: Impacts of El Niño and La Nina on Climate and Society , 2001 .

[42]  John Derber,et al.  An Ocean Analysis System for Seasonal to Interannual Climate Studies , 1995 .

[43]  Stanley B. Goldenberg,et al.  Documentation of a highly ENSO‐related sst region in the equatorial pacific: Research note , 1997 .

[44]  C. F. Ropelewski,et al.  Precipitation Patterns Associated with the High Index Phase of the Southern Oscillation , 1989 .